llmcontroller

Self-hosted AI infrastructure for GGUF inference.
Benchmarking, analytics, orchestration, and runtime control.
NVIDIA • AMD • Linux • Windows

[llmcontroller @llmcontroller]

💥 Coming soon: LLM Controller — the dashboard that gives you control of large language models.
🧠 Self-hosted
⚙️ Real-time stats
📊 Token analytics
🎨 Modern UI

Say goodbye to rented brains. Say hello to your own LLM.

🔗 [https://t.co/eTTh45GW6B]

#SelfHostedAI #LLM #opensource #homelab #localAI #ChatGPTAlternative

[llmcontroller @llmcontroller]

RC1 evolved more than expected!! RC Final is 11 😅

[llmcontroller @llmcontroller]

We have reached Release Candidate on LLM Controller CE. GitHub dropping soon 👀

[llmcontroller @llmcontroller]

14 months of building. Almost ready for GitHub. #inference #LLM #opensource

[llmcontroller @llmcontroller]

@sama @grok give me Samas blog post as a TLDR

[llmcontroller @llmcontroller]

@RealProductGirl https://t.co/K09KT9pP7A

[Nick Dodd @nickdodd]

Damn I spent like $200 on GPU server hours today testing @llmcontroller 😅 plus I got some time from @HotAisle and implemented AMD support! I can't wait to release this software!

[llmcontroller @llmcontroller]

We are so close to Community Edition release! Final polish happening this week! #CommunityAI #AIforeveryone

[llmcontroller @llmcontroller]

@nickdodd @grok

[llmcontroller @llmcontroller]

2025 was big. The vision of LLM Controller was not only born but brought to life bigger and better than we could have imagined!

We didn't make it to public release because it's not perfect yet, our goal is perfection and it will come in 2026! #SelfHostedAI

[llmcontroller @llmcontroller]

We got tired of black-box AI dashboards.

So we built LLM Controller:
• Local inference
• Full GPU visibility
• Real-time tokens/sec
• No cloud dependency

#LocalAI

[Nick Dodd @nickdodd]

@llmcontroller benchmarks are fully functional now, run automatically for newly added models to the local repo, give insight to speed, response quality, expected response length, avg tokens/s and more! I also added support for GPT-OSS and it runs extremely fast! Over 120tps! #LLM

[llmcontroller @llmcontroller]

This is a small sneak peak at LLM Controller!

[Nick Dodd @nickdodd]

I'm working on auto-benchmarking an entire folder of LLM models. So far so good. Once a benchmark is done you can review all of the responses and see all of the stats of the run. New models only auto benchmarked on next run. Change settings to rerun entirely! @llmcontroller #LLM

[Nick Dodd @nickdodd]

I'm working on auto-benchmarking an entire folder of LLM models. So far so good. Once a benchmark is done you can review all of the responses and see all of the stats of the run. New models only auto benchmarked on next run. Change settings to rerun entirely! @llmcontroller #LLM

[llmcontroller @llmcontroller]

Another big day of app updates! We've started closed Beta testing! 🚀

We're also implementing a full benchmark suite into the next internal version!

Big things to come! Stay tuned!

[llmcontroller @llmcontroller]

#LocalAI #LLM #SelfHosted

[llmcontroller @llmcontroller]

Massive day for LLM Controller 🚀
UI polish, smarter workflows, cleaner rendering, and big under-the-hood improvements.
This thing is rapidly turning into a real local LLM control plane.
More soon. 👀🔥

[llmcontroller @llmcontroller]

GPT-OSS will be added to the available models on LLM Controller today!

[llmcontroller @llmcontroller]

@sama Made with GPT-5?

[llmcontroller @llmcontroller]

@sama We are also launching new exciting stuff over the next couple of months 😉

[llmcontroller @llmcontroller]

@nothing @grok @nothing Phone (3) would be an awesome upgrade for my Nothing Phone (2) 😎